From e103e74f0ddff6fb4387020b53471b6e6ace6bd2 Mon Sep 17 00:00:00 2001
From: Greg Way <gregory.way@gmail.com>
Date: Thu, 27 Sep 2018 20:18:13 -0400
Subject: [PATCH] Hotfix Adding Cell Line Analysis (#83)

* add biopython to conda env

* update path to ras_genes.csv

* remove old path to classifiers archive

* add CCLE predictions to analysis
---
 environment.yml                         |  2 ++
 ras_analysis.sh                         | 10 ++++++++++
 scripts/ras_cell_line_predictions.ipynb |  4 ++--
 scripts/viz/ras_summary_figures.R       |  2 +-
 4 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/environment.yml b/environment.yml
index a79a6a8..2dc8936 100644
--- a/environment.yml
+++ b/environment.yml
@@ -1,5 +1,6 @@
 name: pancancer-classifier
 channels:
+- conda-forge
 - bioconda
 dependencies:
 - conda-forge::python=3.6.5
@@ -25,3 +26,4 @@ dependencies:
 - conda-forge::r-ggrepel 0.8.0
 - conda-forge::r-cowplot=0.9.3
 - conda-forge::r-hmisc=4.0_3
+- bioconda::biopython=1.70
diff --git a/ras_analysis.sh b/ras_analysis.sh
index 8e22e7d..2a691f4 100755
--- a/ras_analysis.sh
+++ b/ras_analysis.sh
@@ -182,7 +182,17 @@ python scripts/pancancer_classifier.py --genes 'KRAS,HRAS,NRAS' \
 ###############
 # Step 8. Plot additional Ras, NF1, and BRAF results
 ###############
+# Plot Ras pathway heatmaps
 python scripts/ras_count_heatmaps.py
+
+# Visualize CCLE predictions
+jupyter nbconvert --to=script \
+        --FilesWriter.build_directory=scripts \
+        --ExecutePreprocessor.kernel_name=python3 \
+        --ExecutePreprocessor.timeout=100000 \
+        --execute scripts/ras_cell_line_predictions.ipynb
+
+# Plot summary figures
 Rscript --vanilla scripts/viz/ras_summary_figures.R
 Rscript --vanilla scripts/viz/ras_ccle_pharmacology.R
 Rscript --vanilla scripts/viz/ras_benchmarking_figures.R
diff --git a/scripts/ras_cell_line_predictions.ipynb b/scripts/ras_cell_line_predictions.ipynb
index 7b08e0b..48cc937 100644
--- a/scripts/ras_cell_line_predictions.ipynb
+++ b/scripts/ras_cell_line_predictions.ipynb
@@ -2437,7 +2437,7 @@
    ],
    "source": [
     "# Load TCGA PanCanAtlas Core Ras Pathway genes\n",
-    "ras_genes_file = os.path.join('..', 'classifiers', 'RAS', 'ras_genes.csv')\n",
+    "ras_genes_file = os.path.join('..', 'data', 'ras_genes.csv')\n",
     "ras_core_df = pd.read_table(ras_genes_file)\n",
     "ras_core_df.head()"
    ]
@@ -3622,7 +3622,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.5.5"
+   "version": "3.6.5"
   }
  },
  "nbformat": 4,
diff --git a/scripts/viz/ras_summary_figures.R b/scripts/viz/ras_summary_figures.R
index 04b3769..7f3a7fd 100644
--- a/scripts/viz/ras_summary_figures.R
+++ b/scripts/viz/ras_summary_figures.R
@@ -23,7 +23,7 @@ source(file.path("scripts", "util", "pancancer_util.R"))
 
 set.seed(123)
 
-results_folder <- file.path("classifiers_archive", "RAS")
+results_folder <- file.path("classifiers", "RAS")
 results <- parse_summary(file.path(results_folder, "classifier_summary.txt"))
 
 dir.create("figures")